Commit graph

29 commits

Author SHA1 Message Date
5d93fb7ad3
Default to english 2024-03-14 08:43:16 +01:00
56d7406a31
Add logging to diarization 2023-08-23 17:11:47 +02:00
a79f825f66
Fix speaker rendered throughout paragraph
Each sentence or 'segment' in whisper would be preceded by a [speaker]
notation. This commit fixes that to only include the speaker in front
of a larger group (since a new speaker would start a new diarization
group this will always work).
2023-08-23 15:13:04 +02:00
13b5f22953
Fix Path classes wrongly used in string concats 2023-08-23 15:11:58 +02:00
f7b0d48c6f
Add info logging to failable functions 2023-08-23 15:10:41 +02:00
6cccf00df4
Export TxtTranscription from txt output function 2023-08-23 15:09:53 +02:00
f17639a637
Change argument name for transcription 2023-08-23 15:07:48 +02:00
30fd269cd9
Reinsert whisper model transcribing 2023-08-23 13:23:19 +02:00
3246469be2
Use cuda-enabled base image
Since I can't build cuda images locally (without nvidia gpu), we make use
of an existing one. That also means downgrading python to Debian's version
3.10.
2023-08-22 20:35:25 +02:00
56fa2eca25
Activate venv without poetry 2023-08-22 20:34:12 +02:00
64123a29e0
Reinsert diarization dependencies 2023-08-22 14:27:52 +02:00
48095a1dc9
Fix Dockerfile cmd 2023-08-22 10:55:51 +02:00
66ad116802
REGRESSION: Restrict handler to basic file handling 2023-08-22 10:32:17 +02:00
b28ba0c4d9
Execute docker cmds within poetry environment 2023-08-22 10:10:41 +02:00
6e4ed87813
Make Dockerfile layers more efficient 2023-08-22 10:02:08 +02:00
431c4bea2a
Explicitly set arguments for final docker stage 2023-08-21 18:52:18 +02:00
2b8033dfb2
Copy over README to stop poetry complaining 2023-08-21 18:51:49 +02:00
5b463a6f79
Only run light-the-torch when installed in env 2023-08-21 18:51:23 +02:00
44e8021378
Fix poetry version not given in semver 2023-08-20 14:36:30 +02:00
cb4b633c05
Create modelling processes 2023-08-20 14:29:36 +02:00
6cf1da6ea2
Delete model file 2023-08-20 13:29:32 +02:00
7e91b7a1a2
Create simple loaders infrastructure
Loaders for models, necessary ffmpeg binaries and input files.
2023-08-20 13:29:13 +02:00
540128bc97
Add additional dependencies 2023-08-20 13:28:42 +02:00
6e72776fbf
Install torch models in Dockerfile 2023-08-20 13:28:31 +02:00
79a788692f
Ignore runpod test input files 2023-08-20 12:33:49 +02:00
1451ed705b
Add Dockerfile build 2023-08-20 12:31:49 +02:00
1972dc3525
Add python dependencies 2023-08-20 12:31:33 +02:00
f76213bace
Rename to verbanote 2023-08-20 12:31:14 +02:00
be779f0aca
initial commit 2023-08-20 11:44:03 +02:00